图像增强(IA)、数据预处理

太过爱你忘了你带给我的痛 2023-07-02 13:24 1阅读 0赞

# ![20191009191333910.png][] #

# [日萌社][Link 1] #

[人工智能AI：Keras PyTorch MXNet TensorFlow PaddlePaddle 深度学习实战（不定时更新）][AI_Keras PyTorch MXNet TensorFlow PaddlePaddle]

--------------------

[tensorflow 2.0 画出model网络模型的拓扑图][tensorflow 2.0 _model]

[tensorflow 2.0 的回调函数callbacks（TensorBoard、ModelCheckpoint）][tensorflow 2.0 _callbacks_TensorBoard_ModelCheckpoint]

[TensorBoard视觉化网络每层权重分布、视觉化网络层结构][TensorBoard]

[MSE(均方误差)、MAE(平均绝对误差)][MSE_MAE]

[from\_logits][from_logits]

[CE(Cross-Entropy)、BCE(Binary Cross-Entropy 即Sigmoid+CE)、CCE(Categorical Cross-Entropy 即Softmax+CE)][CE_Cross-Entropy_BCE_Binary Cross-Entropy _Sigmoid_CE_CCE_Categorical Cross-Entropy _Softmax_CE]

[对连续值/离散值进行预处理的两种方式（标准化/one-hot化）、反标准化/逆标准化、字符串预处理][one-hot]

[损失函数loss、指标函数metrics][loss_metrics]

[激活函数 Sigmoid、Tanh、ReLU、Softmax][Sigmoid_Tanh_ReLU_Softmax]

[Batch Normalization][]

[反向传播、梯度下降、学习率、优化器optimizers（GD、SGD、Momentum、AdaGrad、Adam）][optimizers_GD_SGD_Momentum_AdaGrad_Adam]

[权重初始化对于激活函数的选择：随机常态分布/随机正态分布初始化(标准差为0.01或1)、Xavier(Glorot)初始化、He初始化][0.01_1_Xavier_Glorot_He]

[图像增强(IA)、数据预处理][IA]

[混淆矩阵tf.math.confusion\_matrix][tf.math.confusion_matrix]

[使用预训练网络训练的两种方式：Keras Applications、TensorFlow Hub][Keras Applications_TensorFlow Hub]

--------------------

import tensorflow as tf
    
    def parse_aug_fn(dataset):
        """
        Image Augmentation(影像增強) function
        """
        x = tf.cast(dataset['image'], tf.float32) / 255.  # 影像標準化
        x = flip(x)  # 隨機水平翻轉
        # 觸發顏色轉換機率50%
        x = tf.cond(tf.random.uniform([], 0, 1) > 0.5, lambda: color(x), lambda: x)
        # 觸發影像旋轉機率0.25%
        x = tf.cond(tf.random.uniform([], 0, 1) > 0.75, lambda: rotate(x), lambda: x)
        # 觸發影像縮放機率50%
        x = tf.cond(tf.random.uniform([], 0, 1) > 0.5, lambda: zoom(x), lambda: x)
        # 將輸出標籤轉乘One-hot編碼
        y = tf.one_hot(dataset['label'], 10)
        return x, y
    
    
    def parse_fn(dataset):
        x = tf.cast(dataset['image'], tf.float32) / 255.  # 影像標準化
        # 將輸出標籤轉乘One-hot編碼
        y = tf.one_hot(dataset['label'], 10)
        return x, y
    
    
    def flip(x):
        """
        flip image(翻轉影像)
        """
        x = tf.image.random_flip_left_right(x)  # 隨機左右翻轉影像
        return x
    
    
    def color(x):
        """
         Color change(改變顏色)
        """
        x = tf.image.random_hue(x, 0.08)  # 隨機調整影像色調
        x = tf.image.random_saturation(x, 0.6, 1.6)  # 隨機調整影像飽和度
        x = tf.image.random_brightness(x, 0.05)  # 隨機調整影像亮度
        x = tf.image.random_contrast(x, 0.7, 1.3)  # 隨機調整影像對比度
        return x
    
    
    def rotate(x):
        """
        Rotation image(影像旋轉)
        """
        # 隨機選轉n次(通過minval和maxval設定n的範圍)，每次選轉90度
        x = tf.image.rot90(x,tf.random.uniform(shape=[],minval=1,maxval=4,dtype=tf.int32))
        return x
    
    
    def zoom(x, scale_min=0.6, scale_max=1.4):
        """
        Zoom Image(影像縮放)
        """
        h, w, c = x.shape
        scale = tf.random.uniform([], scale_min, scale_max)  # 隨機縮放比例
        sh = h * scale  # 縮放後影像長度
        sw = w * scale  # 縮放後影像寬度
        x = tf.image.resize(x, (sh, sw))  # 影像縮放
        x = tf.image.resize_with_crop_or_pad(x, h, w)  # 影像裁減和填補
        return x

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70][]![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 1][]

[20191009191333910.png]: https://img-blog.csdnimg.cn/20191009191333910.png
[Link 1]: http://www.rimengshe.com/
[AI_Keras PyTorch MXNet TensorFlow PaddlePaddle]: https://blog.csdn.net/zimiao552147572/article/details/88867161
[tensorflow 2.0 _model]: https://azusa.blog.csdn.net/article/details/104092884
[tensorflow 2.0 _callbacks_TensorBoard_ModelCheckpoint]: https://azusa.blog.csdn.net/article/details/104100362
[TensorBoard]: https://azusa.blog.csdn.net/article/details/104132982
[MSE_MAE]: https://azusa.blog.csdn.net/article/details/104094483
[from_logits]: https://azusa.blog.csdn.net/article/details/104095436
[CE_Cross-Entropy_BCE_Binary Cross-Entropy _Sigmoid_CE_CCE_Categorical Cross-Entropy _Softmax_CE]: https://azusa.blog.csdn.net/article/details/104095759
[one-hot]: https://azusa.blog.csdn.net/article/details/104099669
[loss_metrics]: https://azusa.blog.csdn.net/article/details/104099800
[Sigmoid_Tanh_ReLU_Softmax]: https://azusa.blog.csdn.net/article/details/104108314
[Batch Normalization]: https://azusa.blog.csdn.net/article/details/104115714
[optimizers_GD_SGD_Momentum_AdaGrad_Adam]: https://azusa.blog.csdn.net/article/details/104106429
[0.01_1_Xavier_Glorot_He]: https://azusa.blog.csdn.net/article/details/104110857
[IA]: https://azusa.blog.csdn.net/article/details/104114416
[tf.math.confusion_matrix]: https://azusa.blog.csdn.net/article/details/104125078
[Keras Applications_TensorFlow Hub]: https://azusa.blog.csdn.net/article/details/104134536
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70]: https://img-blog.csdnimg.cn/20200130132306304.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0_size_16_color_FFFFFF_t_70 1]: https://img-blog.csdnimg.cn/20200130132311976.png?x-oss-process=image/watermark,type_ZmFuZ3poZW5naGVpdGk,shadow_10,text_aHR0cHM6Ly9henVzYS5ibG9nLmNzZG4ubmV0,size_16,color_FFFFFF,t_70